_ Git - youtube-dl/blob - youtube_dl/extractor/vice.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import time
   6 import hashlib
   7 import json
   8 import random
   9
  10 from .adobepass import AdobePassIE
  11 from .youtube import YoutubeIE
  12 from .common import InfoExtractor
  13 from ..compat import (
  14     compat_HTTPError,
  15     compat_str,
  16 )
  17 from ..utils import (
  18     ExtractorError,
  19     int_or_none,
  20     parse_age_limit,
  21     str_or_none,
  22     try_get,
  23 )
  24
  25
  26 class ViceIE(AdobePassIE):
  27     IE_NAME = 'vice'
  28     _VALID_URL = r'https?://(?:(?:video|vms)\.vice|(?:www\.)?viceland)\.com/(?P<locale>[^/]+)/(?:video/[^/]+|embed)/(?P<id>[\da-f]+)'
  29     _TESTS = [{
  30         'url': 'https://video.vice.com/en_us/video/pet-cremator/58c69e38a55424f1227dc3f7',
  31         'info_dict': {
  32             'id': '5e647f0125e145c9aef2069412c0cbde',
  33             'ext': 'mp4',
  34             'title': '10 Questions You Always Wanted To Ask: Pet Cremator',
  35             'description': 'md5:fe856caacf61fe0e74fab15ce2b07ca5',
  36             'uploader': 'vice',
  37             'uploader_id': '57a204088cb727dec794c67b',
  38             'timestamp': 1489664942,
  39             'upload_date': '20170316',
  40             'age_limit': 14,
  41         },
  42         'params': {
  43             # m3u8 download
  44             'skip_download': True,
  45         },
  46         'add_ie': ['UplynkPreplay'],
  47     }, {
  48         # geo restricted to US
  49         'url': 'https://video.vice.com/en_us/video/the-signal-from-tolva/5816510690b70e6c5fd39a56',
  50         'info_dict': {
  51             'id': '930c0ad1f47141cc955087eecaddb0e2',
  52             'ext': 'mp4',
  53             'uploader': 'waypoint',
  54             'title': 'The Signal From Tölva',
  55             'description': 'md5:3927e3c79f9e8094606a2b3c5b5e55d5',
  56             'uploader_id': '57f7d621e05ca860fa9ccaf9',
  57             'timestamp': 1477941983,
  58             'upload_date': '20161031',
  59         },
  60         'params': {
  61             # m3u8 download
  62             'skip_download': True,
  63         },
  64         'add_ie': ['UplynkPreplay'],
  65     }, {
  66         'url': 'https://video.vice.com/alps/video/ulfs-wien-beruchtigste-grafitti-crew-part-1/581b12b60a0e1f4c0fb6ea2f',
  67         'info_dict': {
  68             'id': '581b12b60a0e1f4c0fb6ea2f',
  69             'ext': 'mp4',
  70             'title': 'ULFs - Wien berüchtigste Grafitti Crew - Part 1',
  71             'description': '<p>Zwischen Hinterzimmer-Tattoos und U-Bahnschächten erzählen uns die Ulfs, wie es ist, "süchtig nach Sachbeschädigung" zu sein.</p>',
  72             'uploader': 'VICE',
  73             'uploader_id': '57a204088cb727dec794c67b',
  74             'timestamp': 1485368119,
  75             'upload_date': '20170125',
  76             'age_limit': 14,
  77         },
  78         'params': {
  79             # AES-encrypted m3u8
  80             'skip_download': True,
  81             'proxy': '127.0.0.1:8118',
  82         },
  83         'add_ie': ['UplynkPreplay'],
  84     }, {
  85         'url': 'https://video.vice.com/en_us/video/pizza-show-trailer/56d8c9a54d286ed92f7f30e4',
  86         'only_matching': True,
  87     }, {
  88         'url': 'https://video.vice.com/en_us/embed/57f41d3556a0a80f54726060',
  89         'only_matching': True,
  90     }, {
  91         'url': 'https://vms.vice.com/en_us/video/preplay/58c69e38a55424f1227dc3f7',
  92         'only_matching': True,
  93     }, {
  94         'url': 'https://www.viceland.com/en_us/video/thursday-march-1-2018/5a8f2d7ff1cdb332dd446ec1',
  95         'only_matching': True,
  96     }]
  97
  98     @staticmethod
  99     def _extract_urls(webpage):
 100         return re.findall(
 101             r'<iframe\b[^>]+\bsrc=["\']((?:https?:)?//video\.vice\.com/[^/]+/embed/[\da-f]+)',
 102             webpage)
 103
 104     @staticmethod
 105     def _extract_url(webpage):
 106         urls = ViceIE._extract_urls(webpage)
 107         return urls[0] if urls else None
 108
 109     def _real_extract(self, url):
 110         locale, video_id = re.match(self._VALID_URL, url).groups()
 111
 112         webpage = self._download_webpage(
 113             'https://video.vice.com/%s/embed/%s' % (locale, video_id),
 114             video_id)
 115
 116         video = self._parse_json(
 117             self._search_regex(
 118                 r'PREFETCH_DATA\s*=\s*({.+?})\s*;\s*\n', webpage,
 119                 'app state'), video_id)['video']
 120         video_id = video.get('vms_id') or video.get('id') or video_id
 121         title = video['title']
 122         is_locked = video.get('locked')
 123         rating = video.get('rating')
 124         thumbnail = video.get('thumbnail_url')
 125         duration = int_or_none(video.get('duration'))
 126         series = try_get(
 127             video, lambda x: x['episode']['season']['show']['title'],
 128             compat_str)
 129         episode_number = try_get(
 130             video, lambda x: x['episode']['episode_number'])
 131         season_number = try_get(
 132             video, lambda x: x['episode']['season']['season_number'])
 133         uploader = None
 134
 135         query = {}
 136         if is_locked:
 137             resource = self._get_mvpd_resource(
 138                 'VICELAND', title, video_id, rating)
 139             query['tvetoken'] = self._extract_mvpd_auth(
 140                 url, video_id, 'VICELAND', resource)
 141
 142         # signature generation algorithm is reverse engineered from signatureGenerator in
 143         # webpack:///../shared/~/vice-player/dist/js/vice-player.js in
 144         # https://www.viceland.com/assets/common/js/web.vendor.bundle.js
 145         # new JS is located here https://vice-web-statics-cdn.vice.com/vice-player/player-embed.js
 146         exp = int(time.time()) + 1440
 147
 148         query.update({
 149             'exp': exp,
 150             'sign': hashlib.sha512(('%s:GET:%d' % (video_id, exp)).encode()).hexdigest(),
 151             '_ad_blocked': None,
 152             '_ad_unit': '',
 153             '_debug': '',
 154             'platform': 'desktop',
 155             'rn': random.randint(10000, 100000),
 156             'fbprebidtoken': '',
 157         })
 158
 159         try:
 160             preplay = self._download_json(
 161                 'https://vms.vice.com/%s/video/preplay/%s' % (locale, video_id),
 162                 video_id, query=query)
 163         except ExtractorError as e:
 164             if isinstance(e.cause, compat_HTTPError) and e.cause.code in (400, 401):
 165                 error = json.loads(e.cause.read().decode())
 166                 error_message = error.get('error_description') or error['details']
 167                 raise ExtractorError('%s said: %s' % (
 168                     self.IE_NAME, error_message), expected=True)
 169             raise
 170
 171         video_data = preplay['video']
 172         base = video_data['base']
 173         uplynk_preplay_url = preplay['preplayURL']
 174         episode = video_data.get('episode', {})
 175         channel = video_data.get('channel', {})
 176
 177         subtitles = {}
 178         cc_url = preplay.get('ccURL')
 179         if cc_url:
 180             subtitles['en'] = [{
 181                 'url': cc_url,
 182             }]
 183
 184         return {
 185             '_type': 'url_transparent',
 186             'url': uplynk_preplay_url,
 187             'id': video_id,
 188             'title': title,
 189             'description': base.get('body') or base.get('display_body'),
 190             'thumbnail': thumbnail,
 191             'duration': int_or_none(video_data.get('video_duration')) or duration,
 192             'timestamp': int_or_none(video_data.get('created_at'), 1000),
 193             'age_limit': parse_age_limit(video_data.get('video_rating')),
 194             'series': video_data.get('show_title') or series,
 195             'episode_number': int_or_none(episode.get('episode_number') or episode_number),
 196             'episode_id': str_or_none(episode.get('id') or video_data.get('episode_id')),
 197             'season_number': int_or_none(season_number),
 198             'season_id': str_or_none(episode.get('season_id')),
 199             'uploader': channel.get('base', {}).get('title') or channel.get('name') or uploader,
 200             'uploader_id': str_or_none(channel.get('id')),
 201             'subtitles': subtitles,
 202             'ie_key': 'UplynkPreplay',
 203         }
 204
 205
 206 class ViceShowIE(InfoExtractor):
 207     IE_NAME = 'vice:show'
 208     _VALID_URL = r'https?://(?:.+?\.)?vice\.com/(?:[^/]+/)?show/(?P<id>[^/?#&]+)'
 209
 210     _TEST = {
 211         'url': 'https://munchies.vice.com/en/show/fuck-thats-delicious-2',
 212         'info_dict': {
 213             'id': 'fuck-thats-delicious-2',
 214             'title': "Fuck, That's Delicious",
 215             'description': 'Follow the culinary adventures of rapper Action Bronson during his ongoing world tour.',
 216         },
 217         'playlist_count': 17,
 218     }
 219
 220     def _real_extract(self, url):
 221         show_id = self._match_id(url)
 222         webpage = self._download_webpage(url, show_id)
 223
 224         entries = [
 225             self.url_result(video_url, ViceIE.ie_key())
 226             for video_url, _ in re.findall(
 227                 r'<h2[^>]+class="article-title"[^>]+data-id="\d+"[^>]*>\s*<a[^>]+href="(%s.*?)"'
 228                 % ViceIE._VALID_URL, webpage)]
 229
 230         title = self._search_regex(
 231             r'<title>(.+?)</title>', webpage, 'title', default=None)
 232         if title:
 233             title = re.sub(r'(.+)\s*\|\s*.+$', r'\1', title).strip()
 234         description = self._html_search_meta(
 235             'description', webpage, 'description')
 236
 237         return self.playlist_result(entries, show_id, title, description)
 238
 239
 240 class ViceArticleIE(InfoExtractor):
 241     IE_NAME = 'vice:article'
 242     _VALID_URL = r'https://www\.vice\.com/[^/]+/article/(?P<id>[^?#]+)'
 243
 244     _TESTS = [{
 245         'url': 'https://www.vice.com/en_us/article/on-set-with-the-woman-making-mormon-porn-in-utah',
 246         'info_dict': {
 247             'id': '41eae2a47b174a1398357cec55f1f6fc',
 248             'ext': 'mp4',
 249             'title': 'Mormon War on Porn ',
 250             'description': 'md5:6394a8398506581d0346b9ab89093fef',
 251             'uploader': 'vice',
 252             'uploader_id': '57a204088cb727dec794c67b',
 253             'timestamp': 1491883129,
 254             'upload_date': '20170411',
 255             'age_limit': 17,
 256         },
 257         'params': {
 258             # AES-encrypted m3u8
 259             'skip_download': True,
 260         },
 261         'add_ie': ['UplynkPreplay'],
 262     }, {
 263         'url': 'https://www.vice.com/en_us/article/how-to-hack-a-car',
 264         'md5': '7fe8ebc4fa3323efafc127b82bd821d9',
 265         'info_dict': {
 266             'id': '3jstaBeXgAs',
 267             'ext': 'mp4',
 268             'title': 'How to Hack a Car: Phreaked Out (Episode 2)',
 269             'description': 'md5:ee95453f7ff495db8efe14ae8bf56f30',
 270             'uploader': 'Motherboard',
 271             'uploader_id': 'MotherboardTV',
 272             'upload_date': '20140529',
 273         },
 274         'add_ie': ['Youtube'],
 275     }, {
 276         'url': 'https://www.vice.com/en_us/article/znm9dx/karley-sciortino-slutever-reloaded',
 277         'md5': 'a7ecf64ee4fa19b916c16f4b56184ae2',
 278         'info_dict': {
 279             'id': 'e2ed435eb67e43efb66e6ef9a6930a88',
 280             'ext': 'mp4',
 281             'title': "Making The World's First Male Sex Doll",
 282             'description': 'md5:916078ef0e032d76343116208b6cc2c4',
 283             'uploader': 'vice',
 284             'uploader_id': '57a204088cb727dec794c67b',
 285             'timestamp': 1476919911,
 286             'upload_date': '20161019',
 287             'age_limit': 17,
 288         },
 289         'params': {
 290             'skip_download': True,
 291         },
 292         'add_ie': [ViceIE.ie_key()],
 293     }, {
 294         'url': 'https://www.vice.com/en_us/article/cowboy-capitalists-part-1',
 295         'only_matching': True,
 296     }, {
 297         'url': 'https://www.vice.com/ru/article/big-night-out-ibiza-clive-martin-229',
 298         'only_matching': True,
 299     }]
 300
 301     def _real_extract(self, url):
 302         display_id = self._match_id(url)
 303
 304         webpage = self._download_webpage(url, display_id)
 305
 306         prefetch_data = self._parse_json(self._search_regex(
 307             r'__APP_STATE\s*=\s*({.+?})(?:\s*\|\|\s*{}\s*)?;\s*\n',
 308             webpage, 'app state'), display_id)['pageData']
 309         body = prefetch_data['body']
 310
 311         def _url_res(video_url, ie_key):
 312             return {
 313                 '_type': 'url_transparent',
 314                 'url': video_url,
 315                 'display_id': display_id,
 316                 'ie_key': ie_key,
 317             }
 318
 319         vice_url = ViceIE._extract_url(webpage)
 320         if vice_url:
 321             return _url_res(vice_url, ViceIE.ie_key())
 322
 323         embed_code = self._search_regex(
 324             r'embedCode=([^&\'"]+)', body,
 325             'ooyala embed code', default=None)
 326         if embed_code:
 327             return _url_res('ooyala:%s' % embed_code, 'Ooyala')
 328
 329         youtube_url = YoutubeIE._extract_url(body)
 330         if youtube_url:
 331             return _url_res(youtube_url, YoutubeIE.ie_key())
 332
 333         video_url = self._html_search_regex(
 334             r'data-video-url="([^"]+)"',
 335             prefetch_data['embed_code'], 'video URL')
 336
 337         return _url_res(video_url, ViceIE.ie_key())